what's wrong with this GainBoosted OPA

[wjxcom]

Hi hr_rezaee: thanx a lot!!

I had already read the paper and found the equation (3), but in this paper, there have the other equation (4) who give the "safe" area of the Unity-Gain Frequency of the aux opa, so I make the Unity-Gain Frequency of the aux opa larger than the main OPA.

regard!!

 

[hr_rezaee]

Hi
Yes, I think you said truth.
you have a second pole near first pole in output
all your problem is it.
try another thing
replace all your CMFBs with ideal then report simulation results.
regards

 

[wjxcom]

Hi hr_rezaee: do you mean that in the MAIN AMP, the second pole is near the first pole? I will replace all my CMFBs with ideal, and modify the MAIN AMP.

 

[hr_rezaee]

Hi

Hi hr_rezaee: do you mean that in the MAIN AMP, the second pole is near the first pole?

I mean the Poles of gain boosted. (outd)
You have second pole near Dominant pole.
so, You have a very ad PM.
for example try to reduce dominant pole (just increase CL of opamp)
you will see all thing is good.
I think your CMFB make another pole near dominant pole.
you must find source of sominant pole.

I will replace all my CMFBs with ideal, and modify the MAIN AMP.

I'm not sure but please try it.
regards

 

[wjxcom]

Hi hr_rezaee: I have already modify the MAIN OPA, i.e. modify the M=70 of M3 and M4 to M=160, and modify the M=60 of M9 and M10 to M=100, so the second pole of the main opa is pushed away, so the PM become 43, the gain become 84dB, and the GBW become 305MHz, the power is 19mW.

At the same time, I replace the CMFBs with ideal, and I found that the OPA with the ideal CMFB has the simulation result of the OPA with the actual CMFBs.

the modified circuit can be find at the attachment.

Thanx!!

 

[hr_rezaee]

ok
thanks
I will see it as soon as possible.
regards

 

[wjxcom]

Hi hr_rezaee: would you give me some advice again? I think I can improve this GainBoosted OPA!

thanx!!

 

[hr_rezaee]

Hi
sorry, I'm busy.
check pole of folding node.
maybe your second pole is it.
drain of M9 and M10 in main amp.
regards

 

[wjxcom]

Hi hr_rezaee: I'm sorry to disturb you! I have check the main amp, and the second is at the drain of M9 and M10, so I increase the current of M7 M8 M9 and M10 to make the second pole larger.

Unfortunately, though the second pole can be made larger, at the same, the gain of the opa is lower, only 84dB.

thanx!!!

 

[hr_rezaee]

Hi
Unfortunately, these are the drawbacks of folded cascode.
you can reduce size of input transisors of auxp and main amp.
but yes, you will lose gain.
another way is to increase load.
it depends on your application and specs.
It was very usefull for me.
rregards.

 

[wjxcom]

Hi hr_rezaee: I think I should improve the gain of the aum amp, it is unacceptable for me to lose gain. In fact, the gain at the specs is 90dB, PM is 65.
Thanx you again to help me.
regards.